Open WebUI

Description

Open WebUI is a self-hosted AI platform for running AI on your own terms. It connects to any model, local or cloud, including Ollama and OpenAI-compatible APIs as well as Anthropic and vLLM, and gives you a provider-agnostic interface where your models, conversations, and tools live in one place. It is extensible, feature-rich, user-friendly, and built to run entirely offline, so your data stays exactly where it belongs.

The platform supports voice, vision, retrieval, generation, and search out of the box, and can be extended with Python through plugins, pipelines, and custom workflows. A community of nearly half a million members shares prompts, models, tools, functions, and discussions that anyone can browse, install, and contribute to. It installs in about a minute via Docker, pip, uv, or a desktop app with no account required, and pairs with Open WebUI Computer, an agent harness and runtime that works on the real files, terminal, git, and services on your own machine.

Built for scale from a laptop to institutional deployments, Open WebUI supports SSO, RBAC, audit logs, data residency, and on-premises or air-gapped deployment for regulated industries, providing the technical foundation for frameworks like SOC 2, HIPAA, GDPR, FedRAMP, and ISO 27001. It is free to use as-is, sustained by sponsors and optional enterprise licenses that add white-labeling and dedicated support, and is used by organizations including Samsung Semiconductor, Astellas Pharma, and a 35,000-user university deployment.

Open WebUI's Core Features

• Self-hosted platform that runs entirely offline

• Provider-agnostic connection to local and cloud models (Ollama, OpenAI, Anthropic, vLLM)

• Voice, vision, retrieval, generation, and search built in

• Python extensibility via plugins, pipelines, and custom workflows

• Community library of prompts, models, tools, and functions to install and share

• One-command install via Docker, pip, uv, or desktop app, no account needed

• Open WebUI Computer agent harness that acts on your real machine

• Enterprise controls: SSO, RBAC, audit logs, data residency, and air-gapped deployment

How to use Open WebUI?

1. 1

   Install it: Run Open WebUI via Docker, pip, uv, or the desktop app in about a minute, no account required.

2. 2

   Connect a provider: Point it at Ollama, OpenAI, Anthropic, vLLM, or any compatible model, local or cloud.

3. 3

   Extend and customize: Add plugins, tools, and Python workflows, or install community-built components.

4. 4

   Scale up: Turn on SSO, RBAC, audit logs, and on-premises or air-gapped deployment as your needs grow.

Open WebUI's Use Cases

• Private self-hosted AI
• Unified multi-model interface
• Extending AI with code
• Regulated enterprise deployment
• Real-machine agents

FAQ from Open WebUI

• What is Open WebUI?

  Open WebUI is a self-hosted, extensible AI platform that connects to any local or cloud model and runs entirely offline, giving you a provider-agnostic interface you control.

• Which models can it connect to?

  It supports Ollama and OpenAI-compatible APIs, as well as Anthropic, vLLM, and anything compatible, both local and cloud-based.

• How do I install it?

  You can install it in about a minute via Docker, pip, uv, or a native desktop app, with no account required to run it.

• Is it suitable for enterprises?

  Yes. It supports SSO, RBAC, audit logs, data residency, and on-premises or air-gapped deployment for regulated industries, and is used by organizations like Samsung Semiconductor and Astellas Pharma.

• What is Open WebUI Computer?

  It is an agent harness and runtime for your real machine that works on the actual files, branches, services, and tools on that computer, with approvals and plan mode, running locally on hardware you own.

• How much does it cost?

  Open WebUI is free to use as-is for everyone, sustained by sponsors and optional enterprise licenses that add white-labeling and dedicated support.